Computer Rating System Prediction Results for College Football (NCAA IA)

2014 Season Totals

Through 2014-08-31
Rank System Pct. Correct Against Spread Absolute Error Bias Mean Square Error games suw sul atsw atsl
1Sagarin0.769230.4102616.0185-2.2938399.326393091623
2Sagarin Golden Mean0.769230.4102616.0185-2.2938399.326393091623
3Sagarin Predictor0.769230.4102616.0185-2.2938399.326393091623
4Sagarin Elo Score0.769230.4102616.0185-2.2938399.326393091623
5Dunkel Index0.769230.5128216.3574-2.9210424.292393092019
6Billingsley0.743590.5897416.2185-3.6113402.7683929102316
7Billingsley+0.743590.6410316.1038-3.0105406.7153929102514
8CPA Retro0.743590.5641016.0169-1.0579391.6653929102217
9Directorofinformation0.743590.5384615.1977-1.9044366.9523929102118
10Super List0.717950.5526316.7697-1.0472446.7543928112117
11Pigskin Index0.717950.4444416.3333-2.0769411.6153928111620
12ARGH Power Ratings0.717950.5263215.7115-2.1218363.7713928112018
13CPA Rankings0.717950.5384615.6564-1.0123382.1003928112118
14Line (midweek)0.7179515.5769-1.2949372.083392811
15Thompson CAL0.717950.5641015.4744-1.3923365.2483928112217
16Pi-Ratings Mean0.717950.6410315.3718-2.0897357.4343928112514
17Computer Adjusted Line0.717950.5555615.2308-1.4615357.2313928112016
18Thompson ATS0.717950.5641015.1769-2.3308353.7003928112217
19Line (updated)0.717950.5757615.1154-1.2692352.7123928111914
20System Median0.692310.4324316.4628-2.4556406.5723927121621
21Ashby AccuRatings0.692310.4444416.5126-2.4613406.5093927121620
22Payne Power Ratings0.692310.4615416.9010-2.3938457.3193927121821
23MDS Model0.692310.5384617.1610-4.2067422.9933927122118
24System Average0.692310.5000016.3915-2.7936403.1493927121919
25Thompson Average0.692310.4359016.2049-2.0192397.9643927121722
26Massey Consensus0.692310.4871816.1013-1.9823415.2343927121920
27Dokter Entropy0.692310.3421116.0454-1.5546397.0153927121325
28Laz Index0.692310.5641015.5854-2.5444380.6753927122217
29Atomic Football0.692310.4615415.4269-2.0459359.5583927121821
30Dave Congrove0.666670.5128217.0015-3.4846439.3473926132019
31Regression-Based Analys0.666670.2894718.0769-1.0513505.7693926131127
32Marsee0.666670.4359018.5897-2.5897523.2053926131722
33Randal Horobik0.666670.3947418.6410-3.7436484.3333926131523
34Pi-Rate Ratings0.666670.5384616.4077-0.6846415.8283926132118
35Howell0.666670.6052616.1156-2.3715408.7773926132315
36PI-Rate Bias0.666670.5384616.1000-0.8949405.9763926132118
37The Power Rank0.666670.5128215.6846-3.2692358.0333926132019
38Line (opening)0.666670.4687515.5897-1.3333370.5903926131517
39Fremeau FEI0.666670.6216215.5641-1.1538369.4623926132314
40Edward Kambour0.648650.4324317.1138-3.1916462.2953724131621
41Sportrends0.647060.4848517.7500-5.9265485.6253422121617
42Covers.com0.641030.6052616.6803-4.9095433.5553925142315
43Keeper0.641030.4102616.9213-1.6787454.1733925141623
44Massey Ratings0.641030.4324316.4359-2.4359405.5133925141621
45Brent Craig0.641030.3333317.1813-1.1921451.6373925141326
46Beck Elo0.641030.3076917.7490-3.8910478.7533925141227
47Laffaye RWP0.641030.5384617.9403-6.1787472.9063925142118
48Compughter Ratings0.641030.4871815.8382-2.1100416.0713925141920
49NutShell Sports0.621620.4594619.6622-4.5541590.7363723141720
50Catherwood Ratings0.615380.4324318.8718-3.5385538.4623924151621
51DP Dwiggins0.615380.3513518.8718-2.1538524.6673924151324
52PerformanZ Ratings0.615380.4359018.4690-4.8285525.7663924151722
53Moore Power Ratings0.594590.2973018.4732-2.7608520.1193722151126
54Donchess Inference0.594590.3783818.8284-3.7305510.6623722151423
55Laffaye XWP0.589740.4102620.0462-1.5026620.4083923161623
56Lee Burdorf0.589740.3589719.3795-4.8564513.9653923161425
57Born Power Index0.589740.2820519.2808-2.6346561.3433923161128
58Daniel Curry Index0.589740.3076918.9667-3.7359542.2573923161227
59Stat Fox0.589740.3421117.7695-3.0510484.0793923161325
60Stephen Kerns0.500000.5135119.4658-5.3605541.4433819191918
61Tempo Free Gridiron0.487180.5384618.8205-7.3333522.8723919202118

* This system does not make predictions.  I make predictions for this
  system by translating it to a new scale that allows for making predictions.



Retrodictive records are found by taking the ratings from the current week
and applying them to the entire season to date.

The ideal system would be one that has the highest correct game decisions,
has the smallest mean error(deviation from the actual game result), and has
a bias of zero.

Mean Error = average[abs(prediction-actual)]

      Bias = agerage(prediction - actual)

      Std. = Standard Deviation of individual game biases